Applied Corpus Linguistics最新文献

With multimodal visual texts becoming increasingly important, especially online, there is a need for an effective tool to analyze these texts. In this paper, we present the Multimodal Corpus Analysis Tool which uses sets of categories, such as dominant colors and object location, to analyze a corpus of online advertisements. The 12 categories are based on both multimodal theory and advertising theory, and the tool produces frequency counts of codes and likelihood of co-occurrence of two codes using observed-to-expected ratio. We illustrate the use of the tool through three case studies. First, examining the left and right positioning of verbal content and images in horizontal adverts, we find that there is a general preference for language left – picture right adverts, a finding potentially questioning currently accepted theories of information value in multimodal texts. Second, examining color schemes in adverts, we find that most use of dominant colors reflects accepted meanings in advice on multimodal text design, except for blue which is associated with storge love. For brightness, colorfulness and contrast, there are differences between adverts for everyday products and adverts for services and expensive products. Third, focusing on the call to action phrases in the adverts, there are differences in the use of the two most common forms (shop now and learn more) in adverts for products and services. These cases illustrate the potential benefits of developing and using multimodal corpus tools.

Marking errors in L2 learner performance, though useful in both a didactic and academic sense, is a challenging process, one usually performed manually when involving learner corpora. This is because errors are largely latent phenomena whose manual identification and description involve a significant degree of judgment on the side of human annotators. The purpose of the paper is to discuss and demonstrate the implications of the two stages of the decision-making process that is manual error coding, error location and error description, for measuring inter-annotator agreement as a marker of quality of annotation. The crux of the study is in the proposal that inter-annotator agreement on error location and on error description should be considered and reported separately rather than, as is common, together as a single measurement. The case study, grounded in a high-stakes exam context and typified using an established error taxonomy, demonstrates the method behind the proposal and showcases its usefulness in real-world settings.

Beyond the context of cross-cultural pragmatics, it has been claimed that race and Rasse are not equivalent terms. The current paper seeks to establish the first known program of study to shed light on the extent to which race and Rasse differ. Corpora of US and German newspapers are used in this paper to sample mainstream race talk in the respective countries, and top collocates are analyzed along with a selection of examples. What is seen is that, while the historical context of slavery and the civil rights movement in the US and Nazism in Germany do seem to be reflected in the respective corpora, it is not so clear that the underlying meaning of race and Rasse are as different as some have claimed.

This study describes the creation and analysis of a small multimodal corpus of British news articles about obesity, where tags were assigned to images in the articles using the automatic tagger Google Cloud Vision. In order to illustrate the potential for analysis of image tags, the corpus analysis tool WordSmith was used to identify differences between newspapers in the ways that obesity was framed. Three forms of analysis were carried out – the first simply compared keywords across the newspapers, the second examined key visual tags and their collocates associated with each newspaper, while the third incorporated a combined analysis of words and image tags. The three analyses produced complementary findings, indicating the value in using Google Cloud Vision in creating and analysing multimodal corpora. The paper ends by reflecting on the method undertaken, while considering how additional research could improve our understanding of image tagging.

Understanding the reception of public health messages in public-facing communications is of key importance to health agencies in managing crises, pandemics, and other health threats. Established public health communications strategies including self-efficacy messaging, fear appeals, and moralising messaging were all used during the Coronavirus pandemic. We explore the reception of public health messages to understand the efficacy of these established messaging strategies in the COVID-19 context. Taking a community-focussed approach, we combine a corpus linguistic analysis with methods of wider engagement, namely, a public survey and interactions with a Public Involvement Panel to analyse this type of real-world public health discourse.

Our findings indicate that effective health messaging content provides manageable instructions, which inspire public confidence that following the guidance is worthwhile. Messaging that appeals to the audience's morals or fears in order to provide a rationale for compliance can be polarising and divisive, producing a strongly negative emotional response from the public and potentially undermining social cohesion. Provenance of the messaging alongside text-external political factors also have an influence on messaging uptake. In addition, our findings highlight key differences in messaging uptake by audience age, which demonstrates the importance of tailored communications and the need to seek public feedback to test the efficacy of messaging with the relevant demographics. Our study illustrates the value of corpus linguistics to public health agencies and health communications professionals, and we share our recommendations for improving the public health messaging both in the context of the ongoing pandemic and for future novel and re-emerging infectious disease outbreaks.

The present study examines the relationship between the use of epistemic stance expressions (i.e., hedges and boosters) and task type, L2 proficiency, and individual authorial style in 1,773 essays representing three different kinds of tasks (complaint, letter, and opinion) written by 591 Finnish L2 English speakers on four different levels of proficiency (CEFR levels B1-C2). The results of the study show that the frequency of both hedges and boosters is mainly governed by task type, as the opinion tex contained a higher number of both hedges and boosters than the other tasks examined. Proficiency-related patterns were, nevertheless, also observed, as it was shown that in the complaint and opinion tasks, the frequency of both hedges and boosters tends to increase with proficiency, while in the letter task, the frequency of both ED types shows signs of decrease. Individual authorial style was shown to play a very limited role in the frequency of EDs in the data, but the results also suggest that the influence of authorial style may be greater with respect to boosters than it is with hedges.

This short ‘methods’ article compares results for six different corpus search techniques for identifying person-first language (e.g. person/people with obesity, person/people with mental illness) and identity-first language (e.g. obese person/people, mentally ill person/people) in a corpus. This distinction is relevant across a range of health contexts, including but not limited to obesity, diabetes, or mental illness. Consequently, there is considerable interest in corpus linguistics and beyond in identifying the frequency of such language in large corpora. However, there is no consensus regarding the specific corpus search techniques to be used for this purpose. This article therefore offers a relevant methodological contribution, based on a trial of six different search techniques. Results from each technique are compared with respect to four different parameters: raw frequency, proportional usage, number of types identified (a proxy for ‘recall’) and false positives (a proxy for ‘precision’). This comparison in turn provides a basis for recommendations for future corpus linguistic studies of person- and identity-first language. The corpus that we use for this trial is a 16.4 million word corpus with newspaper articles containing the word obesity or obese. However, the findings should be relevant to other kinds of identity where similar syntactic structures are at play for expressing identity-first and person-first language.

While much language teaching research has extolled the advantages of using corpora in the language learning classroom, uptake of corpora by language teachers remains low (Poole, 2020; Charles, 2020). Frankenberg-Garcia (2012a) reports that this is partly due to teachers having little time to prepare lessons with corpora given their busy teaching schedules. However, we also argue it is because teachers are not exposed to corpora sufficiently in pre-service teacher training programmes. This paper outlines a training framework for in-service EFL practitioners which will enable them to become corpus literate over a period of weeks. The framework utilises observational learning, and we gauge trainees’ corpus literacy development based on their reflections and experiences of teaching with corpora in class. Similar to previous corpus training research, teachers reported an overwhelmingly positive experience of learning to use a corpus for classroom materials design. Unlike previous trainings with corpora, teachers commented that using corpora did not require more time in lesson preparation when compared with traditional methods. Recommendations for future iterations of the training programme include the incorporation of additional corpora, tools and user interfaces.